Most people who say AI gives them generic answers are not entirely wrong.

They are simply missing half of the equation.

If you type, “Write me a marketing strategy,” you have given the system almost nothing to think with. No market. No customer. No budget. No objective. No constraints. The generic answer that follows is not mysterious.

The professionals getting real value from ChatGPT are learning to brief it the way they would brief a capable colleague: context first, outcome second, constraints clearly stated.

So this is not a list of magic sentences. These are reusable starting points. Replace the brackets with your reality, add relevant documents or examples where appropriate, and challenge the first answer instead of treating it as final.

1. Turn a Messy Idea Into a Clear Plan

I am trying to achieve [goal]. Here is everything currently in my head: [paste notes]. Organise this into a practical plan. Separate what matters now from what can wait, identify missing information, and give me the next three actions in order.

2. Prepare for a Difficult Meeting

I have a meeting with [person/team] about [issue]. My desired outcome is [outcome]. Their likely concerns are [concerns]. Help me prepare: key points, questions I should ask, objections I may hear, and responses that are calm and evidence-based.

3. Make an Email Sound Like You

Rewrite this email so it sounds clear, warm and professional without corporate filler. Keep my meaning intact. Remove unnecessary words and anything that sounds exaggerated or AI-generated: [email].

4. Summarise a Long Document for Decisions

Read the attached/pasted material as if I am the person responsible for making a decision. Give me: the five facts that matter, risks, unresolved questions, deadlines, and what you recommend I investigate before acting.

5. Find the Weakness in Your Idea

Here is my idea: [idea]. Do not sell it back to me. Stress-test it. Identify assumptions, reasons customers may not care, operational problems, competitors or alternatives, and the cheapest way I can test the riskiest assumption.

6. Learn Something Without the Jargon

Teach me [topic] from first principles. Assume I am intelligent but new to the subject. Use one practical analogy, then explain the real terminology. After each major section, give me one question that checks whether I actually understood it.

7. Create a 30-Day Learning Roadmap

I want to become competent in [skill] in 30 days. I can spend [time] per day. My current level is [level]. Build a daily plan focused on practical ability, not passive consumption. Include mini-projects and a way to measure progress.

8. Improve Your CV Bullet Points

Here are my responsibilities and achievements from [role]: [details]. Turn them into concise CV bullets focused on actions and outcomes. Do not invent metrics. Where a number would strengthen a bullet, mark what information I should supply.

9. Prepare for an Interview

I am interviewing for [role]. Here is the job description: [paste]. Based on it, identify the capabilities they are most likely testing. Give me realistic interview questions and help me build STAR examples from my experience: [experience].

10. Research a Company Before Applying

I am considering [company/role]. Using the information I provide and current sources where available, help me understand the business model, customers, competitors, recent priorities, culture signals and questions I should ask before deciding whether it fits me.

11. Turn Meeting Notes Into Accountability

Convert these notes into an action register with: action, owner, deadline, dependency and unresolved question. Do not create owners or deadlines that are not stated; flag missing ones: [notes].

12. Write a Better Proposal

I am proposing [service/solution] to [client type]. Their problem is [problem]. Their desired outcome is [outcome]. Help me structure a proposal around their business case, scope, deliverables, timeline, assumptions and next step. Avoid hype.

13. Analyse Customer Feedback

Here is customer feedback from [source]: [feedback]. Group it into recurring themes, separate symptoms from root problems, show contradictory opinions, and rank the issues by likely impact and frequency.

14. Make a Decision With Trade-offs

I need to choose between [A] and [B]. My priorities are [priorities] and constraints are [constraints]. Build a decision framework, show the trade-offs, identify what would make each option the wrong choice, then recommend based only on my stated priorities.

15. Create Content From Real Experience

I want to write about this experience: [story]. Interview me first. Ask five questions that uncover tension, specific details, what changed my mind and what another professional could learn. Do not draft the post until I answer.

16. Repurpose One Strong Idea

Here is my original article/post: [content]. Extract the strongest single idea and adapt it into a LinkedIn post, a short video concept and an email opening. Each should feel native to its format, not like shortened copies.

17. Audit a Process

Here is how we currently handle [process]: [steps]. Look for duplicated work, delays, manual handoffs, unclear ownership and tasks that could be automated. Redesign the process but preserve any step required for quality, compliance or human judgement.

18. Build an AI Automation Opportunity List

My role/business involves these recurring tasks: [tasks]. Classify them into: automate, AI-assisted, human-led, and do not automate. Explain why and rank the first three experiments by time saved versus implementation risk.

19. Analyse a Spreadsheet or Dataset

I will provide a dataset about [subject]. Before analysing it, tell me what each field appears to represent and flag data-quality problems. Then identify meaningful patterns, anomalies and questions the data cannot answer. Do not confuse correlation with causation.

20. Explain a Technical Error

I am getting this error: [error]. Context: [system/code/actions]. Explain what it means in plain English, list the most likely causes in order, and give me one diagnostic step at a time so I can tell you the result before we continue.

21. Improve a Presentation

Here is my presentation outline: [outline]. The audience is [audience] and I need them to [desired action]. Find where the story loses momentum, remove repetition, identify missing evidence and restructure it into a persuasive narrative.

22. Negotiate Without Sounding Aggressive

I need to negotiate [salary/price/scope/deadline]. Current position: [details]. Desired position: [details]. Help me prepare a respectful negotiation based on value and trade-offs. Give me language for the opening, likely pushback and possible concessions.

23. Plan Your Week Around Outcomes

These are everything I think I need to do this week: [list]. My three most important outcomes are [outcomes]. Build a realistic weekly plan, challenge tasks that do not support those outcomes, and leave buffer for unexpected work.

24. Ask AI to Critique Its Own Answer

Review the answer you just gave me as a skeptical expert. Identify weak assumptions, generic statements, missing evidence and anything that sounds confident without justification. Then produce a stronger version addressing those weaknesses.

25. Create Your Own Reusable AI Brief

Help me create a reusable instruction template for [recurring task]. Interview me about audience, objective, tone, inputs, constraints, examples, quality standard and what a bad result looks like. Then turn my answers into a compact prompt I can reuse.

The Prompt After the Prompt

One habit will improve almost every AI conversation: do not stop at the first answer.

Ask what assumptions it made. Ask what information would change the recommendation. Ask it to remove clichés. Give it a real example. Tell it what you dislike. Make it defend a conclusion.

Good AI use is less like searching Google and more like directing a conversation.

The quality comes from iteration.

A Simple Rule to Remember

Context + clear outcome + constraints + examples + iteration usually beats a clever one-line prompt.

Final Thoughts

Prompt libraries are useful, but they can also create the illusion that there is a secret phrase separating ordinary users from experts.

There isn’t.

The real advantage is learning how to think clearly enough to explain the problem, provide the right context and recognise whether the answer is actually good.

AI can make a capable professional faster. It cannot decide what deserves your judgement.

Bookmark the prompts. But more importantly, learn the habit behind them.